Evapotranspiration estimation with high-resolution aerial imagery over almond orchards using TSEB model

Abstract:

Estimation of evapotranspiration (ET) is important in agriculture for optimum crop water management. ET can be estimated from satellite imagery at larger spatial scales by using surface energy balance models. However, the spatial and temporal resolutions of these imageries are in the range of meters and days, respectively, which is not adequate for crops having large canopy structure such as Almond. In recent years high spatial-temporal aerial imagery provided by commercial companies has become more readily available as an alternative to imagery from satellites. This study aims to evaluate and compare the performance of the physically-based Two-Source Energy Balance (TSEB) modeling scheme for estimating ET over Almond orchards using Landsat and aerial imagery. Both the thermal and infrared mosaicked images were obtained by an airplane (Ceres imaging Inc) overflying an Almond orchard near Corning California during the 2020 growing season with a spatial resolution of 0.5 m. The estimated ET values were compared against eddy covariance measurements within the footprint area for both Landsat and aerial imagery. The results showed that the model performed well in predicting ET from both Landsat and UAV imagery. Pixel by pixel comparison of obtained results indicated that the aerial imagery had closer agreement error between measured and up-scaled fluxes compared to Landsat data. This study demonstrated the potential and suitability of high-resolution aerial imagery for estimation of ET and its implementation for precision orchard irrigation management.